GaussianNB distributed results deviate from the scikit-learn results

[ClaudiaComito]

Description

After the indexing update #758 and subsequent naive_bayes.GaussianNB changes, results from the distributed tests deviate enough from the sklearn comparison that tests fail. Those tests have been temporarily commented out in #758, as the problem is not due to the indexing changes.

TODO: clean up the ht.naive_bayes.GaussianNB implementation, clean separation of local (torch) operations from distributed (heat) operations needed.

To Reproduce Steps to reproduce the behavior:

1. Which module/class/function is affected?
2. What are the circumstances under which the bug appears? unit tests
3. What is the exact error message / erroneous behavior?

Expected behavior Distributed ht.naive_bayes.GaussianNB should return the same result as the non-distributed case (and same as sklearn)

Version Info branch 754-getitem-indexing, PR #758

Additional comments

[mrfh92]

Seems to be a good first issue if you already want to start with sth not that trivial.

Reviewed within #1109

[github-actions[bot]]

Branch bugs/771-GaussianNB_distributed_results_deviate_from_the_scikit-learn_results created!